Desarrollo de un modelo por inteligencia artificial con hemodinamia no invasiva para predecir preeclampsia en embarazos de alto riesgo / Artificial Intelligence Modeling with Non-Invasive Hemodynamics to Predict Preeclampsia in High-Risk Pregnancy

RESUMEN Introducción: la preeclampsia (PE) es la principal causa de morbimortalidad materno-fetal en nuestro país. Alteraciones hemodinámicas precoces durante el embarazo podrían predecir la evolución a PE. El machine learning (ML) permite el hallazgo de patrones ocultos que podrían detectar precozmente el desarrollo de PE. Objetivos: desarrollar un árbol de clasificación con variables de hemodinamia no invasiva para predecir precozmente desarrollo de PE. Material y métodos: estudio observacional prospectivo con embarazadas de alto riesgo (n=1155) derivadas del servicio de Obstetricia desde enero 2016 a octubre 2022 para el muestreo de entrenamiento por ML con árbol de clasificación j48. Se seleccionaron 112 embarazadas entre semanas 10 a 16, sin tratamiento farmacológico y que completaron el seguimiento con el término de su embarazo con evento final combinado (PE): preeclampsia, eclampsia y síndrome HELLP. Se evaluaron simultáneamente con cardiografía de impedancia y velocidad de onda del pulso y con monitoreo ambulatorio de presión arterial de 24 hs (MAPA). Resultados: presentaron PE 17 pacientes (15,18%). Se generó un árbol de clasificación predictivo con las siguientes variables: índice de complacencia arterial (ICA), índice cardíaco (IC), índice de trabajo sistólico (ITS), cociente de tiempos eyectivos (CTE), índice de Heather (IH). Se clasificaron correctamente el 93,75%; coeficiente Kappa 0,70, valor predictivo positivo (VPP) 0,94 y negativo (VPN) 0,35. Precisión 0,94, área bajo la curva ROC 0,93. Conclusión: las variables ICA, IC, ITS, CTE e IH predijeron en nuestra muestra el desarrollo de PE con excelente discriminación y precisión, de forma precoz, no invasiva, segura y con bajo costo.

ABSTRACT Background: Preeclampsia (PE) is the main cause of maternal-fetal morbidity and mortality in our country. Early hemodynamic changes during pregnancy could predict progression to PE. Machine learning (ML) enables the discovery of hidden patterns that could early detect PE development. Objectives: The aim of this study was to build a classification tree with non-invasive hemodynamic variables for the early prediction of PE occurrence. Results: Seventeen patients (15.18%) presented PE. A predictive classification tree was generated with arterial compliance index (ACI), cardiac index (CI), cardiac work index (CWI), ejective time ratio (ETR), and Heather index (HI). A total of 93.75% patients were correctly classified (Kappa 0.70, positive predictive value 0.94 and negative predictive value 0.35; accuracy 0.94, and area under the ROC curve 0.93). Conclusion: ACI, CI, CWI, ETR and HI variables predicted the early development of PE in our sample with excellent discrimination and accuracy, non-invasively, safely and at low cost.

Arterial Stiffness: Its Relation with Prediabetes and Metabolic Syndrome and Possible Pathogenesis.

AIMS: To evaluate arterial stiffness indicators in people with prediabetes (PreD) and its possible pathogenesis. MATERIALS AND METHODS: Pulse wave velocity (PWV) was measured in 208 people with FINDRISC ≥ 13 (57 ± 8 years old, 68.7% women) and thereafter divided into those having either normal glucose tolerance (NGT) or PreD. In each subgroup we also identified those with/out insulin resistance (IR) measured by the triglyceride/HDL-c ratio (normal cut off values previously established in our population). Clinical and metabolic data were collected for all participants. PWV was compared between subgroups using independent t test. RESULTS: Women and men had comparable clinical and metabolic characteristics with obesity (BMI ≥ 30) and antihypertensive-statin treatment, almost half with either NGT or PreD. Whereas 48% of NGT people presented IR (abnormally high TG/HDL-c ratio), 52% had PreD. PWV was significantly higher only in those with a complete picture of metabolic syndrome (MS). CONCLUSIONS: Since PWV was significantly impaired in people with a complete picture of MS, clinicians must carefully search for early diagnosis of this condition and prescribe a healthy life-style to prevent development/progression of CVD. This proactive attitude would provide a cost-effective preventive strategy to avoid CVD's negative impact on patients' quality of life and on health systems due to their higher care costs.

Prevalence of isolated nocturnal hypertension according to 2018 European Society of Cardiology and European Society of Hypertension office blood pressure categories.

OBJECTIVES: To estimate the prevalence of isolated nocturnal hypertension (INH) and its relationships with office blood pressure (BP) categories defined by 2018 ESC/ESH guidelines. METHODS: We conducted a prospective cohort study in consecutive patients referred to perform an ambulatory blood pressure monitoring (ABPM) for diagnosis or therapeutic purposes. Office BP measurements and ABPM were performed in the same visit. The cohort was divided according to office BP in optimal, normal, high-normal and hypertension. The prevalence and adjusted risk for combined daytime and nocturnal hypertension and INH were estimated for each category. RESULTS: We evaluated 1344 individuals, 59.3% women (51 ±â€Š14 years old) and 40.7% men (52 ±â€Š15 years old). 61.5% of the individuals had nocturnal hypertension, 12.9% INH and 48.7% combined daytime and nocturnal hypertension. Prevalence of combined daytime and nocturnal hypertension increased through office BP categories (P < 0.001). Conversely, prevalence of INH was lower in individuals with hypertension than in normotensives (7.4 vs. 17.2%, P < 0.001) and similar between nonhypertensive office BP categories, 16.6, 15 and 19.4% for optimal, normal and high-normal BP, respectively (P < 0.399). In individuals with office BP values less than 140/90 mmHg, the prevalence of masked hypertension phenotypes were 8.6, 17.2 and 30.2% for daytime, INH and combined daytime and nocturnal hypertension, respectively. Adjusted risk for combined daytime and nocturnal hypertension increased significantly through office BP categories; conversely, the risk for INH was similar in all nonhypertensive office BP categories. CONCLUSION: Nocturnal hypertension was the more prevalent phenotype of masked hypertension and more than one-third of the individuals with nocturnal hypertension had INH. The risk for INH was not related to nonhypertensive office BP categories.

Office blood pressure values and the necessity of out-of-office measurements in high-risk pregnancies.

OBJECTIVES: To determine if there is an office blood pressure (BP) value below which out-of-office measurements are unnecessary in high-risk pregnant women. METHODS: We conducted a prospective cohort study in women in the second half of high-risk pregnancies. Office BP measurements and ambulatory blood pressure monitoring (ABPM) was performed. The cohort was divided according to quartiles of office BP and in normotension, white-coat hypertension, masked hypertension and sustained hypertension. The risks for preeclampsia/eclampsia for each category were estimated. RESULTS: Three hundred seventy-three women (30 ±â€Š7 years with 32 ±â€Š4 weeks of gestation) were included; 69 women (18.5%) developed preeclampsia/eclampsia. Risk for preeclampsia/eclampsia increased in a stepwise manner through quartiles of systolic office BP (8.8, 13.4, 19.6 and 32.3%, P < 0.001) and diastolic office BP (6.5, 13.7, 19.6 and 34,4%, P < 0.001). OR increased significantly through quartiles of systolic (P = 0.004) and diastolic (P < 0.001) office BP; the significance becomes evident between the second and third quartile, the cut-off point between these was 125/76 mmHg. Prevalence of white-coat and masked hypertension were 3.8 and 24.7%, respectively. Using ABPM, 14/61 office hypertensive women were reclassified as white-coat hypertension but 92/312 normotensive women as masked hypertension. OR for preeclampsia/eclampsia increased significantly in women with masked hypertension. Absolute risk for preeclampsia/eclampsia in women with office BP less than 125/75 mmHg was similar than that in women with normal ABPM, 7.2 and 7.1%, respectively. CONCLUSION: Masked hypertension was a prevalent and high-risk condition. Office BP at least 125/75 mmHg in the second half of gestation seems appropriate to indicate out-of-office measurements in high-risk pregnancies.

Nocturnal hypertension in high-risk mid-pregnancies predict the development of preeclampsia/eclampsia.

OBJECTIVE: The aim of this study was to test if hypertension detected by ambulatory blood pressure monitoring (ABPM) performed at mid-pregnancy, is a useful predictor for preeclampsia/eclampsia (PEEC). METHODS: The study was performed in women coursing high-risk mid-pregnancies. Office blood pressure (BP) was estimated as the mean of three values, taken by a specialized nurse after a 15-min interview, and office hypertension defined as at least 140/90 mmHg. Immediately after, an ABPM was started. Diurnal hypertension was defined as ABPM at least 135/85 mmHg during daily activities, nocturnal hypertension as ABPM at least 120/70 mmHg during night rest. The adjusted risk of PEEC was estimated using logistic regression. RESULTS: Eighty-seven women (mean age 31 ±â€Š7 years) with 23 ±â€Š2 weeks of pregnancy were included. The prevalence of office and ABPM hypertension was 13.8 and 40.2%, respectively. The concordance between both hypertension diagnosis was low (κ = 0.170, P = 0.044). Nocturnal hypertension (35.6%) was more frequent than diurnal hypertension (26.4%). Nocturnal hypertension markedly increased the relative risk of PEEC (OR 5.32, 95% CI 1.48-19.10). The risk of PEEC attributed to diurnal hypertension did not reach statistical significance; and when both, diurnal and nocturnal hypertension were included in the same model, only the second one was a significant predictor (P = 0.012). The relative risk associated with nocturnal hypertension increased for women not taking acetylsalicylic acid (ASA); (OR 11.40, 95% CI 2.35-55.25). CONCLUSION: Nocturnal hypertension at high-risk mid-pregnancy is a frequent condition and a strong predictor for PEEC; the risk doubled for women not taking ASA.

Could self-measured office blood pressure be a hypertension screening tool for limited-resources settings?

Blood pressure (BP) was assessed by patients themselves in recently published trials. Self-measured office blood pressure (SMOBP) seems particularly interesting for limited health resources regions. The aim of our study was to evaluate the relationship between SMOBP values and those estimated by ambulatory blood pressure monitoring (ABPM). Six hundred seventy-seven patients were evaluated using both, SMOBP and ABPM. The differences between SMOBP and daytime ABPM were evaluated with paired "t" test. The correlations among SMOBP and ABPM were estimated using Pearson's r. The accuracy of SMOBP to identify abnormal ABPM was determined using area under ROC curve (AUC). Sensitivity, specificity, and positive and negative predictive values were calculated for different SMOBP cut-points. Using the average of three readings, systolic SMOBP was higher (3.7 (14.2) mmHg, p < 0.001) and diastolic SMOBP lower (1.5 (8.1) mmHg, p < 0.001) than ABPM. Both BP estimates had a significant correlation, r = 0.67 and r = 0.75 (p < 0.01) for systolic and diastolic BP, respectively. Systolic SMOBP predicted systolic abnormal ABPM; the AUC were 0.80 (0.77-0.84) and 0.78 (0.74-0.81) for daytime and 24 h hypertension, respectively. Diastolic SMOBP predicted diastolic hypertension, AUC 0.86 (0.83-0.88) for both daytime and 24 h hypertension. Neither correlations nor AUCs improved significantly using the average of five readings. SMOBP ≥ 160/90 mmHg was highly specific (>95%) to identify individuals with hypertension in the ABPM; SMOBP < 130/80 mmHg reasonably discarded abnormal ABPM. In conclusion, a high proportion of individuals could be classified adequately using SMOBP, reducing the necessity of healthcare resources and supporting its utility for screening purposes.

Nocturnal but not Diurnal Hypertension Is Associated to Insulin Resistance Markers in Subjects With Normal or Mildly Elevated Office Blood Pressure.

OBJECTIVE: The aim was to evaluate the relationships among insulin resistance markers and nocturnal and diurnal hypertension in normotensive or mildly untreated hypertensive adults. METHODS: The study was performed in both female and male adults referred to the Cardiometabolic Unit of the Hospital San Martín, La Plata, Argentina, in order to perform an ambulatory blood pressure measurement (ABPM) for the evaluation of a possible hypertensive disorder. The population was stratified according to their ABPM in: 1-presence or absence of diurnal hypertension and 2-presence or absence of nocturnal hypertension; both conditions were analyzed separately. Fasting plasma insulin (FPI), homeostasis model assessment of insulin resistance (HOMA-IR), and triglycerides (TG)/high-density lipoprotein cholesterol (HDL-C) ratio were used as surrogate markers of insulin resistance and compared among subjects with vs. without diurnal or nocturnal hypertension. RESULTS: One hundred and five patients, 55 women, 47 (11) years old, and 50 men, 44 (16) years old, were included. Diurnal and nocturnal hypertension were found in 60% and 64% of the sample, respectively. There were no significant differences among the levels of insulin resistance markers between individuals with or without diurnal hypertension. In contrast, individuals with nocturnal hypertension were more insulin resistant irrespectively of whether they were evaluated using FPI (P = 0.016), HOMA-IR (P = 0.019), or TG/HDL-C ratio (P = 0.011); FPI differences remained significant after adjustment for sex, age, and obesity indicators (P = 0.032). CONCLUSIONS: Nocturnal but not diurnal hypertension was related to higher levels of 3 insulin resistance markers in normotensive and untreated mildly hypertensive adults; this relationship seems partially independent of obesity.

Risk of cardiovascular disease according to blood pressure categories in an argentinian cohort.

BACKGROUND: Hypertension is a recognized strong risk factor for cardiovascular disease. However, no data was available in our country to quantify the relationship between blood pressure and cardiovascular event. OBJECTIVE: to quantify the risk of cardiovascular events according to blood pressure categories. METHODS: A prospective epidemiological study was conducted in 1526 inhabitants from Rauch City, (Buenos Aires, Argentina) between 1997 and 2012. Subjects were classified into one of these blood-pressure categories: 1-optimal, 2-normal, 3-high-normal, 4-grade 1 hypertension, 5-grade 2 hypertension and 6-grade 3 hypertension. The first CVD event, including unstable angina pectoris, fatal and non-fatal myocardial infarction, myocardial revascularization, and fatal or non-fatal stroke, was defined as the primary endpoint. Multivariable Cox proportional-hazards regression models were used to estimate the relative risk (HR) of CVD according to base-line blood-pressure categories. RESULTS: In 2012, 1124 individuals (73.7% of the baseline sample), 719 women and 405 men (in 1997, aged 45±16 and 46±16 respectively) or their relatives in case of death, could be surveyed again in order to obtain information concerning incident CVD events. Cardiovascular event rates and HR values increased in a stepwise manner across the blood pressure categories (p for trend across categories <0.001 in both sex); however, in subjects aged ≥55 years a j-curve phenomenon was observed, showing the lowest incidence in the high-normal category. In all categories CVD events rates were higher for men. CONCLUSION: This study quantified relationships between BP and CVD starting from high-normal blood pressure in Argentina.

Significance of masked and nocturnal hypertension in normotensive women coursing a high-risk pregnancy.

OBJECTIVE: The aim was to evaluate the prevalence of nocturnal and masked hypertension and the prognostic values of these blood pressure (BP) abnormalities in normotensive women coursing a high-risk pregnancy. METHODS: The study was performed in pregnant women with 20 or more weeks of gestation coursing a high-risk pregnancy, sent to a specialized hypertension department to perform a prospective defined protocol of BP evaluation. Women with office BP at least 140/90 mmHg were excluded. An ambulatory monitoring of BP was performed to identify masked and nocturnal hypertension (defined according to the current guidelines). The adjusted risk for development of preeclampsia/eclampsia (PEEC) was estimated using logistic regression. The ability of SBP and DBP to identify risk of PEEC was estimated using area under the receiver-operating characteristic curves. RESULTS: Eighty-seven women (29 ±â€Š7 years old, 30 ±â€Š5 weeks of pregnancy) were included in this analysis. The prevalence of masked hypertension was 33.3%. Nocturnal hypertension was found in 42.5% of the women. Remarkably, 27.0% of the women with nocturnal hypertension had normal 24-h values according to ambulatory BP monitoring. Twenty-two patients developed PEEC; adjusted relative risks increased with the presence of nocturnal (odds ratio = 4.72, 95% confidence interval 1.25-19.43, P = 0.023) or masked hypertension (odds ratio = 7.81, 95% confidence interval 2.6-22.86, P = 0.001). Nocturnal SBP and DBP had the highest abilities to predict PEEC (area under the curve = 0.77 and 0.80, respectively). CONCLUSION: Masked and nocturnal hypertension are frequent findings in normotensive women coursing a high-risk pregnancy, and their presence implies an increased risk to develop PEEC.

Insulin resistance: The linchpin between prediabetes and cardiovascular disease.

The aim of this study was to test the hypothesis that cardiovascular disease occurs to the greatest extent in persons with prediabetes mellitus who are also insulin resistant. In 2003, 664 non-diabetic women (n = 457) and men (n = 207), aged 52 ± 16 and 53 ± 15 years, were surveyed during a programme for cardiovascular disease prevention. Fasting plasma glucose concentrations defined participants as having normal fasting plasma glucose (fasting plasma glucose <5.6 mmol/L) or prediabetes mellitus (fasting plasma glucose ⩾ 5.6 and <7.0 mmol/L). The tertile of prediabetes mellitus subjects with the highest fasting plasma insulin concentration was classified as insulin resistant. Baseline cardiovascular disease risk factors were accentuated in prediabetes mellitus versus normal fasting glucose, particularly in prediabetes mellitus/insulin resistant. In 2012, 86% of the sample were surveyed again, and the crude incidence for cardiovascular disease was higher in subjects with prediabetes mellitus versus normal fasting glucose (13.7 vs 6.0/100 persons/10 years; age- and sex-adjusted hazard ratio = 1.88, p = 0.052). In prediabetes mellitus, the crude incidences were 22.9 versus 9.6/100 persons/10 years in insulin resistant versus non-insulin resistant persons (age- and sex-adjusted hazard ratio = 2.36, p = 0.040). In conclusion, cardiovascular disease risk was accentuated in prediabetes mellitus/insulin resistant individuals, with a relative risk approximately twice as high compared to prediabetes mellitus/non-insulin resistant subjects.

Use of the plasma triglyceride/high-density lipoprotein cholesterol ratio to identify cardiovascular disease in hypertensive subjects.

This analysis evaluated the hypothesis that the plasma triglyceride (TG)/high-density lipoprotein cholesterol (HDL-C) concentration ratio can help identify patients with essential hypertension who are insulin-resistant, with the cardiovascular disease (CVD) risk profile associated with that defect. Data from a community-based study developed between 2003 and 2012 were used to compare CVD risk factors and outcome. Plasma TG/HDL-C cut-points of 2.5 (women) and 3.5 (men) subdivided normotensive (n = 574) and hypertensive (n = 373) subjects into "high" and "low" risk groups. Metabolic syndrome criteria (MetS) were also used to identify "high" and "low" risk groups. The baseline cardio-metabolic profile was significantly more adverse in 2003 in "high" risk subgroups, irrespective of BP classification or definition of risk (TG/HDL-C ratio vs. MetS criteria). Crude incidence of combined CVD events increased across risk groups, ranging from 1.9 in normotensive-low TG/HDL-C subjects to 19.9 in hypertensive-high TG/HDL-C ratio individuals (P for trends <.001). Adjusted hazard ratios for CVD events also increased with both hypertension and TG/HDL-C. Comparable findings were seen when CVD outcome was predicted by MetS criteria. The TG/HDL-C concentration ratio and the MetS criteria identify to a comparable degree hypertensive subjects who are at greatest cardio-metabolic risk and develop significantly more CVD.

Blood pressure response to a community-based program and long-term cardiovascular outcome.

BACKGROUND: The significance in terms of the cardiovascular outcome of different patterns of blood pressure (BP) response to community-based activities has not been previously studied. The aim of our study was to evaluate the relationship between changes of BP observed during a community-based program and long-term rate of cardiovascular disease (CVD) events. METHODS: A program focused on the prevention of CVDs was implemented in a small city in Argentina over a 6-year period (1997-2003). The program was evaluated using a cohort of 1,526 randomly selected individuals. The population's mean systolic BP (SBP) decreased approximately 5mm Hg, but BP changes were heterogeneous. Eight years after the intervention had stopped, 1,124 individuals of the cohort were surveyed to evaluate incident CVD events. The sample was divided into 3 SBP change categories during the intervention: decrease (< -5mm Hg), no change (-5 to 5mm Hg) and increase (> 5mm Hg). Cox models were used to estimate the hazard ratio (HR) of CVD events for each category adjusted for age, sex, and baseline BP. RESULTS: At the end of the intervention, individuals who had increased their SBP showed an adjusted HR for CVD that was double those whose SBP levels decreased or did not change. An inverse relationship between baseline SBP and ΔSBP was observed: individuals with an increase in BP during the intervention had lower SBP at the beginning of the study. CONCLUSIONS: BP increase during the community-based program was an independent predictor of CVD events.